The proposed research draws together an epidemiological concern for occupational characteristics as risk factors for health and mental health with a sociological concern about status attainment, social mobility and the consequences of mental disorders on the processes. Integrating these two concerns in one study is imperative since without a careful consideration of how individuals select into occupations estimates of the effects of occupational characteristics on mental health are likely to be confounded. Two already collected sets of data, the Dictionary of Occupational Titles ratings of occupational characteristics and the Washington Heights PERI-Risk Factor study will be merged and used in the analysis. These data are readily available and key variables are known to have sound psychometric properties (Dohrenwend et al., 1980; Cain and Treiman, 1981). The Dictionary of Occupational Titles data are objective ratings and provide more detailed measures of occupational position than were available in other studies of mobility. The PERI-Risk Factor data contain information on both risk factors and psychiatric condition for community, first break patients with schizophrenia and major depression and repeat admission patient samples with schizophrenia and major depression. Using these data, analyses will be conducted to examine three specific aims. First, the effects of various types of psychiatric disorder and labeling on downward mobility will be examined using the measures derived from the Dictionary of Occupational Titles ratings as well as more traditional measures of occupational prestige. Second, the effects of occupational characteristics on psychiatric condition will be assessed taking stock, of course, of the selection possibility studied in the first specific aim. Finally by comparing the characteristics of occupations for treated and untreated cases who have been successful in their occupational careers to those who have not been so fortunate, we hope to provide information that will be of use to those concerned with placing former patients in jobs they are likely to be successful at. The social mobility of groups varying in psychiatric condition will be analyzed using LISREL while the case control component testing the effects of occupational characteristics will employ logistic regression.